Upload trainer_state.json
Browse files- trainer_state.json +166 -0
trainer_state.json
ADDED
@@ -0,0 +1,166 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": null,
|
3 |
+
"best_model_checkpoint": null,
|
4 |
+
"epoch": 1.8226342885780673,
|
5 |
+
"global_step": 5000,
|
6 |
+
"is_hyper_param_search": false,
|
7 |
+
"is_local_process_zero": true,
|
8 |
+
"is_world_process_zero": true,
|
9 |
+
"log_history": [
|
10 |
+
{
|
11 |
+
"epoch": 0.18,
|
12 |
+
"learning_rate": 0.00015,
|
13 |
+
"loss": 2.8412,
|
14 |
+
"step": 500
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"epoch": 0.18,
|
18 |
+
"eval_loss": 1.5813477039337158,
|
19 |
+
"eval_phone_accuracy": 0.5792233379941528,
|
20 |
+
"eval_runtime": 11.4324,
|
21 |
+
"eval_samples_per_second": 17.494,
|
22 |
+
"eval_steps_per_second": 4.374,
|
23 |
+
"step": 500
|
24 |
+
},
|
25 |
+
{
|
26 |
+
"epoch": 0.36,
|
27 |
+
"learning_rate": 0.0003,
|
28 |
+
"loss": 1.482,
|
29 |
+
"step": 1000
|
30 |
+
},
|
31 |
+
{
|
32 |
+
"epoch": 0.36,
|
33 |
+
"eval_loss": 0.8674991726875305,
|
34 |
+
"eval_phone_accuracy": 0.7458370408033558,
|
35 |
+
"eval_runtime": 10.836,
|
36 |
+
"eval_samples_per_second": 18.457,
|
37 |
+
"eval_steps_per_second": 4.614,
|
38 |
+
"step": 1000
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 0.55,
|
42 |
+
"learning_rate": 0.00026656263932233615,
|
43 |
+
"loss": 1.0056,
|
44 |
+
"step": 1500
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 0.55,
|
48 |
+
"eval_loss": 0.5736156702041626,
|
49 |
+
"eval_phone_accuracy": 0.8264268463200711,
|
50 |
+
"eval_runtime": 10.9515,
|
51 |
+
"eval_samples_per_second": 18.262,
|
52 |
+
"eval_steps_per_second": 4.566,
|
53 |
+
"step": 1500
|
54 |
+
},
|
55 |
+
{
|
56 |
+
"epoch": 0.73,
|
57 |
+
"learning_rate": 0.0002331252786446723,
|
58 |
+
"loss": 0.7876,
|
59 |
+
"step": 2000
|
60 |
+
},
|
61 |
+
{
|
62 |
+
"epoch": 0.73,
|
63 |
+
"eval_loss": 0.43626296520233154,
|
64 |
+
"eval_phone_accuracy": 0.8651752044404898,
|
65 |
+
"eval_runtime": 11.1035,
|
66 |
+
"eval_samples_per_second": 18.012,
|
67 |
+
"eval_steps_per_second": 4.503,
|
68 |
+
"step": 2000
|
69 |
+
},
|
70 |
+
{
|
71 |
+
"epoch": 0.91,
|
72 |
+
"learning_rate": 0.00019968791796700847,
|
73 |
+
"loss": 0.6745,
|
74 |
+
"step": 2500
|
75 |
+
},
|
76 |
+
{
|
77 |
+
"epoch": 0.91,
|
78 |
+
"eval_loss": 0.3738095760345459,
|
79 |
+
"eval_phone_accuracy": 0.8828545400618618,
|
80 |
+
"eval_runtime": 10.7535,
|
81 |
+
"eval_samples_per_second": 18.599,
|
82 |
+
"eval_steps_per_second": 4.65,
|
83 |
+
"step": 2500
|
84 |
+
},
|
85 |
+
{
|
86 |
+
"epoch": 1.09,
|
87 |
+
"learning_rate": 0.0001662505572893446,
|
88 |
+
"loss": 0.5965,
|
89 |
+
"step": 3000
|
90 |
+
},
|
91 |
+
{
|
92 |
+
"epoch": 1.09,
|
93 |
+
"eval_loss": 0.3428588807582855,
|
94 |
+
"eval_phone_accuracy": 0.8905342993940935,
|
95 |
+
"eval_runtime": 10.807,
|
96 |
+
"eval_samples_per_second": 18.507,
|
97 |
+
"eval_steps_per_second": 4.627,
|
98 |
+
"step": 3000
|
99 |
+
},
|
100 |
+
{
|
101 |
+
"epoch": 1.28,
|
102 |
+
"learning_rate": 0.00013281319661168078,
|
103 |
+
"loss": 0.5429,
|
104 |
+
"step": 3500
|
105 |
+
},
|
106 |
+
{
|
107 |
+
"epoch": 1.28,
|
108 |
+
"eval_loss": 0.29903942346572876,
|
109 |
+
"eval_phone_accuracy": 0.9031926613279099,
|
110 |
+
"eval_runtime": 11.2035,
|
111 |
+
"eval_samples_per_second": 17.852,
|
112 |
+
"eval_steps_per_second": 4.463,
|
113 |
+
"step": 3500
|
114 |
+
},
|
115 |
+
{
|
116 |
+
"epoch": 1.46,
|
117 |
+
"learning_rate": 9.937583593401693e-05,
|
118 |
+
"loss": 0.5022,
|
119 |
+
"step": 4000
|
120 |
+
},
|
121 |
+
{
|
122 |
+
"epoch": 1.46,
|
123 |
+
"eval_loss": 0.2947227954864502,
|
124 |
+
"eval_phone_accuracy": 0.9031185119274607,
|
125 |
+
"eval_runtime": 10.8322,
|
126 |
+
"eval_samples_per_second": 18.464,
|
127 |
+
"eval_steps_per_second": 4.616,
|
128 |
+
"step": 4000
|
129 |
+
},
|
130 |
+
{
|
131 |
+
"epoch": 1.64,
|
132 |
+
"learning_rate": 6.593847525635309e-05,
|
133 |
+
"loss": 0.4767,
|
134 |
+
"step": 4500
|
135 |
+
},
|
136 |
+
{
|
137 |
+
"epoch": 1.64,
|
138 |
+
"eval_loss": 0.2698778808116913,
|
139 |
+
"eval_phone_accuracy": 0.9127261556713698,
|
140 |
+
"eval_runtime": 11.228,
|
141 |
+
"eval_samples_per_second": 17.813,
|
142 |
+
"eval_steps_per_second": 4.453,
|
143 |
+
"step": 4500
|
144 |
+
},
|
145 |
+
{
|
146 |
+
"epoch": 1.82,
|
147 |
+
"learning_rate": 3.2501114578689255e-05,
|
148 |
+
"loss": 0.4565,
|
149 |
+
"step": 5000
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"epoch": 1.82,
|
153 |
+
"eval_loss": 0.2591952979564667,
|
154 |
+
"eval_phone_accuracy": 0.9146540400830473,
|
155 |
+
"eval_runtime": 11.459,
|
156 |
+
"eval_samples_per_second": 17.454,
|
157 |
+
"eval_steps_per_second": 4.363,
|
158 |
+
"step": 5000
|
159 |
+
}
|
160 |
+
],
|
161 |
+
"max_steps": 5486,
|
162 |
+
"num_train_epochs": 2,
|
163 |
+
"total_flos": 1.8035219197147087e+18,
|
164 |
+
"trial_name": null,
|
165 |
+
"trial_params": null
|
166 |
+
}
|